Intentional insulin omission for weight control is a significant problem in the clinical management of type 1 diabetes (T1D), associated with early and severe diabetes-related medical complications and premature death. Up to 40% of young women with T1D misuse insulin to control weight, meeting diagnostic criteria for an eating disorder (ED). Omitting insulin allows individuals to eat and maintain or even lose weight by excreting glucose into the urine; however, it also contributes to a 3 to 14-fold increase in mortality risk. Treatments for T1D patients with EDs are underdeveloped and understudied. There is no clear evidence that current strategies improve metabolic control in this high-risk population, and there has been limited treatment innovation to produce more compelling outcomes. We have collected daily, in-the-moment data on thoughts, feelings, and behaviors that are associated with intentional insulin omission, as well as factors that increase T1D patients' vulnerability to ED symptomatology (5R01 DK-089329-03). These data provide new insights on predictors and correlates of insulin omission that allow us to tailor treatments to thi patient population and reduce insulin misuse in daily life. In the current project, we translate these emerging findings into an innovative treatment strategy that uses a smartphone multimedia application (app) to help individuals recognize their triggers for insulin misuse in-the moment and cope more effectively. This is the first treatment that is based on the systematic study of intentional insulin omission in the natural environment. Because the intervention is based on data that are highly ecologically valid, it is more likely to engage critical treatment targets. This is also the first study to leverage smartphones to overcome psychological barriers to T1D management. Delivering intervention to patients at the time and place when it is most needed may be a more effective and efficient strategy to achieve behavior change, and allow patients to more readily generalize skills to their home environment. In order to more swiftly bring clinical resources to this patient population, we build our innovative treatment tool on the infrastructure of an existing mobile app for EDs, Recovery Record. In Aim 1, we build T1D-specific content into the app based on our empirical evidence of triggers for insulin omission and qualitative feedback from T1D patients. The fully functional app is beta tested with T1D patients who intentionally omit insulin. In Aim 2, we test whether app-supported individual treatment is associated with improved metabolic control and a decrease in ED symptoms and emergency care utilization. We also examine correspondence between frequency of app use and outcomes and calculate attrition. Project deliverables include (1) a mobile app that is accessible to T1D patients that may reduce intentional insulin omission, and (2) procedures for treating insulin misuse to be tested in a larger clinical trial. The long-term goal of this research is to establishan empirically-supported treatment that helps T1D patients reduce life-threatening insulin misuse and preserve longevity and quality of life.